一种改进的K-means聚类彩色图像分割方法  

Improved K-means Clustering for Color Image Segmentation Method

作　　者：刘小丹[1] 牛少敏[1] 

机构地区：[1]辽宁师范大学计算机与信息技术学院,辽宁大连116081

出　　处：《湘潭大学自然科学学报》2012年第2期90-93,共4页Natural Science Journal of Xiangtan University

基　　金：国家自然科学基金项目(60773031);辽宁省高等学校优秀人才支持计划基金项目(200811833)

摘　　要：图像分割是从图像处理到图像分析的关键步骤.图像分割的目的是将图像分割为多个互不重叠且又各具特性的区域,主要应用于图像压缩、目标提取、模式识别等.以往的图像分割技术主要应用于灰度图像,随着计算机技术的进步,彩色图像分割逐渐受到关注.该文在前人对彩色图像分割问题的大量研究成果基础上,提出了一种将K-means聚类、蚁群算法以及分水岭算法相结合的分割方法.本方法有效的克服了聚类数目必须依据先验知识提前设定、最初的聚类中心是随机选取的、聚类的效果好坏依赖于距离判定公式的缺陷.Image segmentation is the key step from image processing to the image analysis. The aim is to put the image segmentation for multiple overlapping and each with different characteristics and the region, mainly used in image compression target extraction pattern recognition, etc. Previous image segmentation technology is mainly used in gray image, along with the rapid development of computer technology, color image segmentation gain more and more attention. Based on previous color image segmentation problem of a lot of the basis of research achievements, this paper proposes a will K-means clustering algorithm,ant colo- ny algorithm and the combination of watershed image segmentation method. The proposed method is effec- tive to overcome the cluster number must be based on prior knowledge set in advance of the initial clustering center is randomly selected cluster effect quality depends on the defects of the judge distance formula.

关 键 词：K—means聚类 彩色图像分割 蚁群算法 分水岭算法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]